“If this is happening to me, an AI expert, what happens to other people?”
This week I published a blog on a subject that I’d been researching and hurriedly included quotes to further strengthen the point I was trying to make. As it transpired, those quotes were bogus and in fact AI hallucinations.
It would seem this is another dirty secret at the heart of the AI industry - AI hallucinations! And it’s a fact not widely known amongst users. Especially those of us using on-line resources for research purposes.
On 22 Apr 2026, Mark Sweney wrote a piece in The Guardian, AI hallucinations found in high-profile Wall Street law firm filing. An elite law firm in New York had to apologise to a federal judge for a string of errors as a result of hallucinations generated by artificial intelligence. The co-head of the firm’s global restructuring group, apologised in a letter to the judge for a string of mistakes, which included inaccurate citations.
On May 24 2026, Tristan Bove wrote a piece for Fortune, AI hallucinations are infiltrating expert work - and entering the permanent body of knowledge. An associate professor at Columbia University’s School of Nursing had grown accustomed to having artificial intelligence tools help polish scientific papers for grammar, formatting, and other details. But a few weeks after submitting his latest research, the academic journal he was due to publish in, came back with questions about a reference. The AI tool Topaz had used inserted a fabricated source into his work. “I felt deeply embarrassed,” Topaz, who leads a team at Columbia developing AI applications in healthcare. “I’m an AI researcher. I know about hallucinations,” he said. “If this is happening to me, an AI expert, what happens to other people?”
Then two days ago, on June 12 2026, Elizabeth Bratton in London and Stephen Foley in New York, wrote for the Financial Times, A KPMG report on how AI is being used by businesses across the world exaggerated adoption of the technology with bogus case studies that appear to have been based on AI hallucinations. The October report, “Redefining excellence in the age of agentic AI”, made numerous false claims about the use of AI by organisations including the Swiss bank UBS, the UK’s National Health Service and the public transit groups Swiss Federal Railways and Transport for London. The inaccuracies were identified as AI hallucinations by the research group GPTZero and verified by the Financial Times. After being alerted to the issue, UBS said it would ask KPMG to remove the false claims, and the Big Four firm on Thursday pulled the report from some of its websites. Hallucinations are a growing problem across the professional services. Elite law firm Sullivan & Cromwell also admitted in April that a filing it submitted in a bankruptcy case contained numerous AI-generated inaccuracies, including misreadings of the US bankruptcy code.
What I’m beginning to grasp is that artificial intelligence is either A) at its core a servile sycophant, or B) overbearingly conceited. Could it be so eager to please that it will take a subject you’re researching and if it can’t find the data to support your enquiry, it will spit back utter nonsense, to bolster your point of view. Or if it is unable to retrieve the information you require from its universal knowledge bank, it will literally make sh*t up, rather than disclose that it's incapable of answering your query. Either way, it’s not a resource in and of itself. You have to dig deeper and read the articles it’s taking its references from, or you’ll come a cropper.
There is nothing more sacred for me than the written word. And what I’m beginning to understand is that if I read it on-line, I must do my due diligence. This space is not sacred - it hallucinates!